Interference management in energy-efficient heterogeneous networks

By the densification of the cellular networks, the performance of the users, who are close to coverage boundaries, significantly degrades due to severe interference. The statistical characteristics of the interfering signal gets complicated by heterogeneity of the base station deployment and employing non-orthogonal multiple access (NOMA). In this thesis, stochastic geometry analysis is presented to reveal the statistical characteristics of the user performance. The interference experienced by the user includes the signals coming from the neighbor base stations with different transmit power capabilities and also from user's own serving base station for NOMA. Coverage probabilities of strict fractional frequency reuse (FFR) and soft frequency reuse (SFR) techniques are derived for NOMA-based heterogeneous network to indicate their performance under such interference. Parameter optimization problems for strict FFR and almost blank sub-frame (ABS) are modelled for maximization of energy efficiency by the capacity derivations of stochastic geometry analysis. A hybrid multiple access method, where NOMA and orthogonal multiple access (OMA) jointly deployed for improving edge user performance, is proposed and compared with conventional NOMA. For uplink NOMA, a decode order optimization problem is presented with its solution for avoidance of intra-cell interference under the user transmit power constraints.